The Document Output Accuracy chart displays the accuracy of the final output, regardless of whether it was created by the machine or a keyer, for fields and table cells that were sampled for Quality Assurance (QA).
For Structured documents, this value is the measure of the output Transcription Accuracy.
For Semi-structured documents, the following applies:
For fields, this value is a combination of Field ID and Transcription Accuracy.
For table cells, it is a combination of Table ID Transcription Accuracy.
Note that, for fields, the chart reports on an occurrence level rather than on a field level.
Under the chart, you can find the Field Accuracy percentage and the Table Cell Accuracy percentage. These percentages reflect the average accuracy over the selected date range.
To see additional performance metrics, hover your cursor over the data point of interest, and a window will display the accuracy percentage for that date, the number of machine-extracted fields and table cells, and the total number of fields and table cells.
Unless QA Sample Rate is 100%, these accuracy numbers are an estimate of what the actual Document Output Accuracy is and will have a margin of error.
Filtering options
The following filtering options are available in the Document Output Accuracy chart:
Date range: by default, these figures are generated with the last week's data, but can be recalculated for a custom date range.
Document type: specify either Structured or Semi-structured forms.
Field type: by default, all field types will be shown, but you can specify Text, Checkboxes, and/or Signatures.
Flow type: by default, all flows will be displayed, but you can specify one or more flows to report instead.
Settings
The Exclude Auto-transcribed Fields toggle allows you to exclude fields that were transcribed by the machine without Supervision, regardless of confidence.
Note that reporting data with excluded auto-transcribed fields is available only from the date of installing or upgrading to v35.